[gnuastro-devel] [bug #56859] Bad result and crash of MakeCatalog and Table with PV parameters in header

[Raul Infante-Sainz]

URL:
  <https://savannah.gnu.org/bugs/?56859>

                 Summary: Bad result and crash of MakeCatalog and Table with
PV parameters in header
                 Project: GNU Astronomy Utilities
            Submitted by: infantesainz
            Submitted on: Wed 04 Sep 2019 12:16:14 PM UTC
                Category: MakeCatalog
                Severity: 3 - Normal
              Item Group: Crash
                  Status: None
                 Privacy: Public
             Assigned to: None
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

I found a no sense result and crash when trying to obtain a catalogue of
objects using a combination of Noisechisel, Segment and MakeCatalog. Step by
step:


I have two images: `sip.fits' and `pv.fits'. The names came from the different
notation for the distortion coefficients (SIP and PV). Both images are the
same, as it can be seen with `DS9' matching WCS.


Then I run the shell script `script' attached in this message. Basically, it
runs NoiseChisel, Segment and finally MakeCatalog over both images.


At the end I get two catalogues with different results. The catalogue from
`sip.fits' is good but the catalogue obtained from `pv.fits' is wrong.


Seems to me that it is reading the WCS information from the PV keywords in the
header in a bad way, and the result are wrong RA-DEC coordinates for the
detected objects.

In addition to this, if I try to use the feature `wcstoimg' with Table, I get
an error in the case of PV but not in the case of SIP. That is, once the
script has been executed I try to do:


asttable sip_cat.fits -cRA,DEC,"arith RA DEC wcstoimg" --wcsfile=sip.fits


It success. But if I do:


asttable pv_cat.fits -cRA,DEC,"arith RA DEC wcstoimg" --wcsfile=pv.fits

It fails with an error:


asttable: gal_wcs_world_to_img: wcss2p ERROR 9: One or more of the world
coordinates were invalid


Here I paste the content of `script':




# Define the name of the input images
i_sip=sip.fits
i_pv=pv.fits



# Run Noisechisel
sip_det=sip_det.fits
pv_det=pv_det.fits

astnoisechisel $i_sip --output=$sip_det
astnoisechisel $i_pv --output=$pv_det



# Run Segment
sip_seg=sip_seg.fits
pv_seg=pv_seg.fits

astsegment $sip_det --output=$sip_seg
astsegment $pv_det --output=$pv_seg



# Run MakeCatalog
sip_cat=sip_cat.fits
pv_cat=pv_cat.fits

astmkcatalog $sip_seg --ra --dec --magnitude --output=$sip_cat
astmkcatalog $pv_seg --ra --dec --magnitude --output=$pv_cat




  



    _______________________________________________________

File Attachments:


-------------------------------------------------------
Date: Wed 04 Sep 2019 12:16:14 PM UTC  Name: script  Size: 531B   By:
infantesainz

<http://savannah.gnu.org/bugs/download.php?file_id=47450>
-------------------------------------------------------
Date: Wed 04 Sep 2019 12:16:14 PM UTC  Name: pv.fits  Size: 638KiB   By:
infantesainz

<http://savannah.gnu.org/bugs/download.php?file_id=47451>
-------------------------------------------------------
Date: Wed 04 Sep 2019 12:16:14 PM UTC  Name: sip.fits  Size: 638KiB   By:
infantesainz

<http://savannah.gnu.org/bugs/download.php?file_id=47452>

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?56859>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/